Software Engineer 5 – Training Platform, AI Platform
Core
Design and build scalable infrastructure for large-scale machine learning model training, fine-tuning, and evaluation workflows.
Role type
Senior IC distributed systems engineer (ML training platform)
Builds
Training platform for foundation models and generative AI
Domain
Cloud computing + Machine Learning
Deliverable
production ML models
Required skills
distributed model training, Kubernetes, Ray clusters, PyTorch, cloud computing (AWS), observability, logging, on-call processes
Preferred skills
Generative AI (foundation model training, fine-tuning, distillation), distributed training parallelism (FSDP, tensor/pipeline), cloud-based AI/ML services (SageMaker, Bedrock, Databricks), performance analysis tools (PyTorch Profiler, NVIDIA Nsight)
Technologies
Kubernetes, Ray, PyTorch, AWS
Responsibilities
Design and build platform infrastructure, libraries, and SDKs for large-scale model training; Diagnose and optimize performance of distributed training jobs (GPU utilization, memory efficiency, fault tolerance); Design APIs and interfaces for ML practitioners and non-experts
Seniority
Senior, hands-on IC
